Skip to content

spec(port): land 25 trios-chat CR-CHAT-04/07 port PRs in one merge (Closes #10096) - #10097

Merged
gHashTag merged 27 commits into
masterfrom
sweep/land-28-port-prs
Oct 17, 2026
Merged

gHashTag merged 27 commits into
masterfrom
sweep/land-28-port-prs

Conversation

@gHashTag

Copy link
Copy Markdown
Owner

Closes #10096

Lands these 25 port PRs, one commit each, original author kept:

Why one PR: each original adds one row to docs/reports/t27b_expectations.json, so every merge on master made the next one conflict again before its ~29 jobs left the Actions queue.

How the ledger was resolved: the rows from both sides were unioned, sorted, kept in master's 2-space layout, with one verdict per path (the resolver stops on a path that has two different verdicts). The ledger diff is +25 rows and nothing else.

Checked on the Railway lab, master 9506315 against this head:

  • tools/dupe_scan.py: the same 3 errors as master (hashes32_equal, is_zero_id, sched_sqrt are already on master), no new one;
  • the diff touches only specs/port/**.t27, .trinity/seals/*.json and the ledger; no foreign code, no secret-shaped added line.

Left out: #9650, #9651 and #9652, because each adds another copy of ids_equal / hashes32_equal.

🤖 Generated with Claude Code

dmitrii-f-t27 and others added 25 commits October 18, 2026 01:41
Landed from #9641 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9660 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9669 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9670 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9678 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9679 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…#8884)

Landed from #9695 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9697 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
… (Refs #8884)

Landed from #9700 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9708 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…8884)

Landed from #9739 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…#8884)

Landed from #9754 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9757 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9794 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
)

Landed from #9802 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9805 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9810 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…(Refs #8887)

Landed from #9811 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
)

Landed from #9813 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…efs #8887)

Landed from #9814 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…8887)

Landed from #9819 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…(Refs #8887)

Landed from #9820 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…8887)

Landed from #9822 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9826 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Landed from #9828 as part of one combined PR (the Actions queue could not
carry 28 single runs while master kept moving under the ledger rows).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-17 19:06:39 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 48
PRs with All Checks Green 2
READY 0
FAILING 48
PENDING 0
NO CHECKS YET 0

These columns do not partition: 0 + 48 + 0 + 0 = 48, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=e769ef021713 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

…het, #2774)

cover_traffic_emission_pattern_randomness_guard.t27 (from #9811) declared
`IntervalObservation`, a name master already defines in
cover_traffic_inter_arrival_autocorrelation_guard.t27. Two definitions of
one type name leave its domain size undetermined, so `tri types ratchet`
went red (ledger 109, observed 110). Rename it EmissionIntervalObservation
in this file only and reseal; the ratchet is CLEAN at 109 again.

Refs #10096

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-17 19:31:11 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 50
PRs with All Checks Green 0
READY 0
FAILING 50
PENDING 0
NO CHECKS YET 0

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=3b53d48ddb52 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

gHashTag added a commit that referenced this pull request Oct 17, 2026
The batch brought four type names that two specs each define, and two
definitions of one name leave its domain size undetermined:
`tri types ratchet` went red, ledger 109 vs observed 113.

- IntervalObservation: the emission pattern randomness guard (#9811)
  repeats master's inter-arrival guard. Renamed EmissionIntervalObservation,
  byte-identical to #10097's fix, so the two batches merge cleanly.
- SizeObservation / SizeDistVerdict: message_size_class_distribution_guard
  repeats cover_traffic_size_distribution_guard. Renamed SizeClassObservation
  / SizeClassDistVerdict.
- TransitionObservation: padding_class_transition_entropy_guard repeats
  padding_length_class_transition_uniformity_guard. Renamed
  EntropyTransitionObservation.

Each renamed spec is resealed. Checked on the Railway lab: types ratchet
CLEAN at 109, dupe_scan finds no new duplicate body.

Refs #8884

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
#10098 landed 20 of this batch's PRs with byte-identical files, so only
docs/reports/t27b_expectations.json conflicted. Resolved as the union of
both sides' rows (sorted, one verdict per path; no path had two verdicts).
Against master this PR now adds 5 specs, their 5 seals and 5 ledger rows
(#9739 #9754 #9757 #9794 #9802).

Refs #10096

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@gHashTag
gHashTag disabled auto-merge October 17, 2026 19:39
@gHashTag
gHashTag enabled auto-merge October 17, 2026 19:39
@gHashTag

Copy link
Copy Markdown
Owner Author

Merged master after #10098 landed (4ab1ed8). #10098 already carried 20 of this batch's PRs with byte-identical files, so only the ledger conflicted, and it was resolved as the row union. Against master, this PR now adds the remaining 5: #9739 #9754 #9757 #9794 #9802 (5 specs, 5 seals, 5 ledger rows). The red t27b-native-ratchet is master's own: emit_kit was fixed by #10103, which arrives with this merge, and rust_convert is tracked in #9637. Auto-merge has been switched from rebase to a merge commit because the branch now contains a merge.

@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-17 19:41:10 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 50
PRs with All Checks Green 0
READY 0
FAILING 50
PENDING 0
NO CHECKS YET 0

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=3b53d48ddb52 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

@gHashTag
gHashTag merged commit 56e8f58 into master Oct 17, 2026
29 of 32 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Land 25 green trios-chat port PRs as one combined PR

2 participants